A METHOD FOR ESTIMATING MEAN AREAL RAINFALL USING MOVING TREND FUNCTIONS OF THE INTENSITIES

This paper investigates the internal spatial distribution of rain rate s inside the rain areas. Moving trend functions are defined for differ ent thresholds tau. They are obtained as conditional mathematical expe ctations of the rain rates above tau, and depend on the distance betwe en the considered point with rain rate higher than tau and the boundar y of the tau-thresholded area, which is the area where the intensity i s above tau. These functions are linked to the dynamics of rainfall pa tterns and are thus named moving trend functions. Such functions are c alculated on a dataset of hourly rainfall fields recorded in 1989 and 1990 in the framework of the Epsat-Niger project. The study area is ab out 12 000 km(2) and is instrumented with a dense gauge network. The s hape of the moving trend functions show that, on average, the rain int ensity increases from the edge to the center of the tau-thresholded ar eas. Thus, the spatial distribution of the rain rates depends on the s hape of the tau-thresholded areas.An estimation algorithm for mean are al rainfall is then proposed using the moving trend functions. This es timation implicitly depends on the shape of the tau-thresholded area. The algorithm is applied to two subareas of 400 and 3600 km(2) of the Niger dataset. The method is also compared with the threshold method o n these two subareas. The two techniques give equivalent results on th e 3600-km(2) area and some small improvements can be observed on the 4 00-km(2) area, especially for low thresholds. However, problems encoun tered in the estimation of the moving trend functions discussed in the paper and due to the finite size of the study area prevent the moving trend function technique to demonstrate its potentiality clearly.